Getting it right the last time
First versions are supposed to be wrong. The question is whether you go back.
Yesterday I sat down to do a quality pass on something I thought was already working. I asked it six realistic questions. One of them it could not answer — not because the answer did not exist, but because I had built the right answer into the wrong place.
That was the fifth thing I found and fixed. The sixth came an hour later.
There is a line I keep returning to: It's not at all important to get it right the first time. It's vitally important to get it right the last time. I used to read that as permission to be sloppy early. I do not read it that way anymore.
The first half is obvious and true. Every first version is wrong. That is not a failure of skill or attention — it is the nature of building something before you fully understand what you are building. You do not know the edge cases until something walks into them. You do not know which question someone will actually ask until they ask it.
The second half is where most people fail. Getting it right the last time requires you to keep going back. It requires a kind of humility that does not come naturally to someone who has just spent three weeks building a thing: the humility to assume that what you cannot yet see is still broken.
The quality pass is not the cleanup after the work. It is the work.
The quiet kind of wrong
What I was fixing yesterday was not crashes. It was the quiet kind of wrong — a gap nobody would notice until they noticed, a guard that protected one path and left another unlatched, a silence that ended something when it should have prompted again. None of it would have started a fire. All of it would have caused the low-grade friction that makes someone stop using a thing without being able to name why.
That is the dangerous failure mode for a solo founder. Not the catastrophic break someone sends you a message about, but the slow erosion of trust from something that is almost right. Almost right is indistinguishable from slightly wrong, and slightly wrong reads as not caring.
When I found number six in that pass, I did not feel bad about it. I fixed it. The mistake would have been stopping at five and calling it done.
You will be wrong the first time. That is fine and expected. What is not optional is going back — until the thing works the way someone other than you actually needs it to.
Getting it right the last time is not a deadline. It is a standard.